Reloading: another interesting phenomenon - I see this regularly... - Page 2

Reloading: another interesting phenomenon - I see this regularly...

This is a discussion on Reloading: another interesting phenomenon - I see this regularly... within the Defensive Rifles & Shotgun Discussion forums, part of the Related Topics category; Well, in any case, it definitely looks like things are improving as the loads go up, and we already know from experience there are a ...

Page 2 of 3 FirstFirst 1 2 3 LastLast
Results 16 to 30 of 32
Like Tree35Likes

Thread: Reloading: another interesting phenomenon - I see this regularly...

  1. #16
    Member Array Bongo Boy's Avatar
    Join Date
    Dec 2008
    Location
    Manitou Springs CO
    Posts
    131
    Well, in any case, it definitely looks like things are improving as the loads go up, and we already know from experience there are a good number of cases where *light* loads, at least, produce erratic results--so I'm saying I agree, not a lot of shooting is probably required to see better results at the higher end.

    What drives me crazy is when the labels have come off my boxes of hand loads, or I've not labelled them at all, I'm at the range and the groups are either noticeable bad or exceptionally good. I guess this is my well-deserved penalty for sloppy work.
    Tangle likes this.
    ďIf I had to live my life again, I'd make the same mistakes, only sooner.Ē T Bankhead
    NRA Benefactor Member

  2. #17
    VIP Member Array ColoradoDiablo's Avatar
    Join Date
    Jul 2016
    Location
    Colorado
    Posts
    2,340
    Quote Originally Posted by CavemanBob View Post
    I shoot a prairie dog with a 22 LR, it makes a little hole in, little hole out.
    I shoot a prairie dog with a 22-250, prairie dog comes apart at the seams. Yes, there's hydrostatic shock effect, velocity matters!
    We need to shoot some together!

    .17 HMR knocks the heck out of a prairie dog.
    CavemanBob likes this.
    U.S. Army, Retired (1986 to 2014)
    Life Member, Veterans of Foreign Wars

  3. #18
    Distinguished Member Array CavemanBob's Avatar
    Join Date
    May 2015
    Location
    Kalirado
    Posts
    1,627
    Quote Originally Posted by ColoradoDiablo View Post
    We need to shoot some together!

    .17 HMR knocks the heck out of a prairie dog.
    We should try to do that, maybe when it's a little warmer out, a bit chilly today.

  4. Remove Advertisements
    DefensiveCarry.com
    Advertisements
     

  5. #19
    VIP Member Array Struckat's Avatar
    Join Date
    Feb 2014
    Location
    Lurking
    Posts
    3,855
    What is SD? As soo as you go acronym I lose all context.
    Tangle likes this.
    Read our government's plan to destroy the Constitution.
    https://sustainabledevelopment.un.or...ormingourworld

  6. #20
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,908
    Quote Originally Posted by Struckat View Post
    What is SD? As soo as you go acronym I lose all context.
    Sorry @Struckat , my bad, I usually do better than that. Typically I say "SD (Standard Deviation)". SD is kind of a quantitative way of expressing how tightly the velocities are about the average velocity.

    For example, consider the following 5 velocities

    2500
    2510
    2520
    2530
    2540

    The average is 2520, the SD is 15.81

    If we change the 2510 to 2505 and the 2530 to 2535, we have

    2500
    2505
    2520
    2535
    2540

    Notice the average is exactly the same, 2520, but the SD has increased to 17.68 - because the numbers aren't as tight around the average.
    Struckat likes this.
    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  7. #21
    VIP Member Array Struckat's Avatar
    Join Date
    Feb 2014
    Location
    Lurking
    Posts
    3,855
    Thank you, I had sectional density mind block, which obviously didnít fit.
    Tangle likes this.
    Read our government's plan to destroy the Constitution.
    https://sustainabledevelopment.un.or...ormingourworld

  8. #22
    VIP Member Array OldVet's Avatar
    Join Date
    Nov 2009
    Location
    The Florida Twilight Zone
    Posts
    32,236
    Quote Originally Posted by Struckat View Post
    Thank you, I had sectional density mind block, which obviously didnít fit.
    Beats a dense mind being sectionally blocked.
    Tangle likes this.
    Retired USAF E-8. Curmudgeon on the loose.
    Lighten up and enjoy life because:
    Paranoia strikes deep, into your life it will creep. It starts when you're always afraid... Buffalo Springfield - For What It's Worth

  9. #23
    VIP Member Array ColoradoDiablo's Avatar
    Join Date
    Jul 2016
    Location
    Colorado
    Posts
    2,340
    Quote Originally Posted by CavemanBob View Post
    We should try to do that, maybe when it's a little warmer out, a bit chilly today.
    I was in Grand Lake yesterday and it was colder today in Boulder County than in Grand Lake yesterday.
    U.S. Army, Retired (1986 to 2014)
    Life Member, Veterans of Foreign Wars

  10. #24
    Senior Moderator
    Array gasmitty's Avatar
    Join Date
    Jun 2007
    Location
    Gilbert, AZ
    Posts
    18,852
    Quote Originally Posted by Tangle View Post
    I've been refining a load for my 10.5" Seekins Precision SBR. So far, the best bullet has been the Nosler 70 gr RDF. I've pretty well focused on Ramshot TAC powder after trying about 8 different powders. Not saying TAC is the best or that I'm done evaluating powders. I've changed my testing and now it's pretty well "standardized" which consists of a 3x10 series (10 powder charges, 3 shots each) in 0.2 grain increments. So I will have to go back and reevaluate some powders. But for now, it's TAC.

    So the setup is the Seekins Precision 10.5" barrel, a Gemtech Trek T suppressor, and an ATN x sight 4K 5-20 digital scope.

    It's pretty easy to understand that guns have "sweet spots", i.e. bullets and powder combos that shoot more accurately and/or consistently than other loads. But, what's a bit harder for me to understand is why the SD's vary so much over the 10 stage ladder. Here's a table of what I shot yesterday:


    What you want to notice here is the large swings in the standard deviation of the velocity (blue curve) for the lower charges, 24.4 - 25.6 gr, and the reduction in variation and continuing decline as the charge weight increases with the higher charges, 25.8 - 26.2 gr. THAT'S WHAT I SEE A LOT! Also, notice that the velocity is increasing with charge in essentially a straight line. That is, as the charge increases, the velocity increases proportionally. Keep in mind, that velocity is the average velocity of the three shots at each charge.
    Tangle, as a data junkie I spent a chunk of time digesting all of this today. Nothing really leaped out at me. But - I really think you may be too focused on standard deviations, especially given that you're applying that calculation to 3-shot strings. By linear regression, the r-squared value (correlation factor) is only 0.39, which suggests virtually no correlation between SD and charge weight, velocity or MOA. I just think a greater sample or population size is required to make applying a standard deviation calculation meaningful. What might be useful is to apply the SD calculation to maybe 10-shot groups once you've narrowed down your load choices to just a few. Right now it appears to be a calculational artifact without much significance.

    In the midst of this I started to wonder if unfilled volume in the cartridge case was a factor. Loading primarily pistol cartridges with a very low-charge weight powder (Titegroup), I'm used to seeing less than half the case filled with powder. So I found the nominal case volume for a 5.56x45, and then the density of the powder you're using, and wow - it looks like there's almost no free volume in the case! Which prompts the next question - is there any chance you're compressing the powder charge? How tightly the powder settles in the case before inserting and seating the bullet might be another variable, although heaven knows how you could measure or control that.

    Just some rambling musings.
    Struckat and BamaT like this.
    Smitty
    AZCDL Life Member
    NRA Patron Member
    NROI Chief Range Officer

  11. #25
    VIP Member Array Struckat's Avatar
    Join Date
    Feb 2014
    Location
    Lurking
    Posts
    3,855
    Quote Originally Posted by OldVet View Post
    Beats a dense mind being sectionally blocked.
    I resemble that remark.
    Read our government's plan to destroy the Constitution.
    https://sustainabledevelopment.un.or...ormingourworld

  12. #26
    VIP Member Array Sticks's Avatar
    Join Date
    Nov 2007
    Location
    Colorado
    Posts
    4,157
    Quote Originally Posted by ColoradoDiablo View Post
    We need to shoot some together!

    .17 HMR knocks the heck out of a prairie dog.
    Quote Originally Posted by CavemanBob View Post
    We should try to do that, maybe when it's a little warmer out, a bit chilly today.
    Heading up to the Pawnee Grasslands Thursday or Friday. If I get my preferred spot, 1.5 MOA steel out to 750. 1 mile east of Bakers Draw. Have some load development to do, and (roo roo) Becky is wanting to go play. Company is welcome if you're inclined.

    Tangle - watching the trend at your 26.2 charge, I would be inclined to push past that at .1gr increments. I think you are coming up on the second sweet spot with velocity, SD/ES and accuracy.
    Sticks

    Grasseater // Grass~eat~er noun, often attributive \ˈgras-ē-tər\
    A person who is incapable of independent thought; a person who is herd animal-like in behavior; one who cannot distinguish between right and wrong; a foolish person.
    See also Sheep

  13. #27
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,908
    Quote Originally Posted by gasmitty View Post
    Tangle, as a data junkie I spent a chunk of time digesting all of this today. Nothing really leaped out at me. But - I really think you may be too focused on standard deviations, especially given that you're applying that calculation to 3-shot strings. By linear regression, the r-squared value (correlation factor) is only 0.39, which suggests virtually no correlation between SD and charge weight, velocity or MOA. I just think a greater sample or population size is required to make applying a standard deviation calculation meaningful. What might be useful is to apply the SD calculation to maybe 10-shot groups once you've narrowed down your load choices to just a few. Right now it appears to be a calculational artifact without much significance.

    In the midst of this I started to wonder if unfilled volume in the cartridge case was a factor. Loading primarily pistol cartridges with a very low-charge weight powder (Titegroup), I'm used to seeing less than half the case filled with powder. So I found the nominal case volume for a 5.56x45, and then the density of the powder you're using, and wow - it looks like there's almost no free volume in the case! Which prompts the next question - is there any chance you're compressing the powder charge? How tightly the powder settles in the case before inserting and seating the bullet might be another variable, although heaven knows how you could measure or control that.

    Just some rambling musings.
    Thanks - good stuff! I do appreciate that! But as they say, however...

    I'm afraid when it comes to reloading, especially load development, we're stuck with low samples. I doubt there are many that would have the inclination, patience, time, and money to shoot statistically relevant numbers of shots per group or charge. Typically, to confirm our scope's zero, we shoot 3 shot groups. To confirm accuracy, some shoot 3, some 5, some 10 shot groups. All are statistically low samples, very low in fact. Consider if we wanted to know the average size of the male shoe. Would 10 samples be statistically sufficient? Goodness no. Plus, if I'm looking at load performance, loading 10 rounds for 10 different charges would be 100 rounds per bullet and/or gun to be tested. It's just not possible for me to do.

    Since we are restricted to low samples by inclination, time and money, we have to relax some of the statistical/math requirements if we want to use some of these tools to help us analyze and quantify results. When the goal is load development, i.e. what bullet/charge best meets our needs, we have to decide how we want to go about that and how we will analyze the results. The most popular methods of load development use 1 - 5 shots per charge, all of which are statistically insufficient from a purely statistical perspective.

    Even though the samples are low for true statistical purposes, some very interesting observations must be recognized. There is much significance in the fact that high and erratic SDs are present ONLY in the low charges, and significantly lower and tighter and more consistent SDs are present ONLY in the higher charges. If we assume that the higher, erratic SDs are inaccurate due to the low samples, why don't we see the same effect for both low and high charges? It would be logical to expect even distribution of variations throughout the charge ranges, and we don't see that.

    That suggests two things. One is that the SDs are indicative and reliable within limits even though the samples are small. And two, it suggests that the data is not necessarily independent and random as required for statistical analysis. If the SDs met the independent and random requirements we would very likely not see large variations in low charges and much tighter variations in higher charges.

    Also, you guys don't get to see what I see so it does put you at a disadvantage. For example, the plot in the OP is not an isolated 3 shot phenomenon. As I mentioned in the OP, I see this a lot. Here's some examples with different guns and loads:

    12/31/2018


    12/30/2018






    This last one is kinda special. I shot ten odd charges, i.e. 24.1, 24.3, etc. one day and went back the next day with the very same loads but shot ten even number charges, i.e. 24.2, 24.4, etc. Then I combined the data into a composite and got the following plot. I also added the second day's even data to see if the data points would land on or off the composite curve - they landed on the composite curve! Although curve fitting is yet another issue and I'll just stick with the data points themselves for simplicity sake. And I could show the "curves" as straight lines and I doubt we would gain any information.



    These are just the plots that I already have uploaded to Imgur for posting purposes; I have more. One thing they have in common is SDs are significantly higher and spread out for lower charges than they are for the higher charges. So again, if there is error due to small sample sizes, we should see it across the charges rather than just one section of charges.

    As for the relationship of SD to accuracy, I see none, i.e. no correlation at all. I've seen some of the lowest SDs produce the largest groups and the highest SDs produce the tightest groups and pretty much everything in between. This came as a surprise to me. I expected the best accuracy to exactly coincide with the best SDs. That theory evaporated quickly.

    I want to address the thought about unfilled volume in the case due to low charge weights. First, that's an excellent point. Second, notice that idea comes from the observation that the SDs are higher and wider for low charge weights than the higher charge weights. Hence the powder is packed more tightly in higher charges and more loosely in the lighter charges. I find it interesting that the SD data suggests this even from the three shot samples!

    My conclusions to this point:
    I think the 3 shot per charge test I'm running is a reasonably reliable indicator of bullet/gun performance at various charges. It's by no means 100%, but then what predictive test is. But I do see very similar results from test to test.

    The data is insufficient to stand on its own so I do compare it to the "collective" of data I have recorded over and over.

    The data is sufficient to select a velocity and/or charge to test in more detail. For example, I'm looking for a consistent SD no higher than 15 fps, and extreme spread of no more than about 30 fps, an accuracy of about 0.75 MOA, and a muzzle energy of about 1050 ft-lbs from a 10.5" - 11.5" gun. That's the muzzle energy I need to retain 600 ft-lbs at 300 yards with a high BC bullet.

    If I'm interested in 500 yards, I need a little tighter velocity variation.

    So once the 3x10 identifies possible loads, I load up 5 rounds, 0.1 gr below, 5 rounds at the selected charge, and 5 rounds 0.1 gr higher. If this gives results that meets the expectations, I load up 30 rounds at the best charge and do a 3-5 round zero at 100 yards, a 5 shot drop test at 300 yards, calculate the BC from the muzzle velocity and drop, set the BC in my ballistics calculator to that BC and adjust the scope for a 300 yard shot. IF it's right, I calculate the drop for 200 yards and adjust the scope for that range and shoot 3-5 shots. They are usually right on the vertical they should be after the BC correction. The remaining rounds - they're just for fun!
    gasmitty likes this.
    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  14. #28
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,908
    Quote Originally Posted by Sticks View Post
    ...Tangle - watching the trend at your 26.2 charge, I would be inclined to push past that at .1gr increments. I think you are coming up on the second sweet spot with velocity, SD/ES and accuracy.
    Thanks @Sticks !

    I think you're right. I would normally reshoot the very same load again as a verification, but I've been thinking I should do exactly what you said. So in order to keep at a 3 x 10, I think I'll drop the two lower charges and add two 0.1 gr charges. That would give me a 26.3 and 26.4 charge.

    I'm almost inclined to go in 0.2 gr increments, but that may be a bit too much. Gotta do some figuring on that.

    Thanks again!
    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  15. #29
    Distinguished Member Array CavemanBob's Avatar
    Join Date
    May 2015
    Location
    Kalirado
    Posts
    1,627
    Quote Originally Posted by Sticks View Post
    Heading up to the Pawnee Grasslands Thursday or Friday. If I get my preferred spot, 1.5 MOA steel out to 750. 1 mile east of Bakers Draw. Have some load development to do, and (roo roo) Becky is wanting to go play. Company is welcome if you're inclined.

    Tangle - watching the trend at your 26.2 charge, I would be inclined to push past that at .1gr increments. I think you are coming up on the second sweet spot with velocity, SD/ES and accuracy.
    I'd be so inclined if not for the fact that I'll be busy on my employer's treadmill.

  16. #30
    VIP Member Array Sticks's Avatar
    Join Date
    Nov 2007
    Location
    Colorado
    Posts
    4,157
    Quote Originally Posted by CavemanBob View Post
    I'd be so inclined if not for the fact that I'll be busy on my employer's treadmill.
    It's flu season. I can forge a DR. note as good as the next guy.
    Tangle and CavemanBob like this.
    Sticks

    Grasseater // Grass~eat~er noun, often attributive \ˈgras-ē-tər\
    A person who is incapable of independent thought; a person who is herd animal-like in behavior; one who cannot distinguish between right and wrong; a foolish person.
    See also Sheep

Page 2 of 3 FirstFirst 1 2 3 LastLast

Sponsored Links

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •